Previous studies have found that a higher white blood cell (WBC) count is associated with incident stroke. There have been few studies examining differential WBC counts in elderly or Asian populations. We studied the association between total and differential WBC counts and incident stroke in an older Asian population.
The Honolulu Heart Program is a prospective population-based study of cardiovascular diseases in Japanese-American men that started in 1965. At exam 4 (1991–93), 3,741 men ages 71–93 years participated, and total and differential WBC counts were measured in 3,569 men using a Coulter counter machine. Data on incident stroke (all strokes [ALL-CVA], thromboembolic [TE-CVA] and hemorrhagic [HEM-CVA]) were available through December 1999 (8 years follow-up) from a comprehensive hospital surveillance system. After excluding 227 subjects with prevalent stroke, 3,342 subjects were divided into quartiles of total WBC, neutrophil (segmented and band), granulocyte (neutrophil, eosinophil and basophil), lymphocyte, and monocyte counts for separate analyses.
Age-adjusted incident ALL-CVA rates increased significantly with total WBC quartiles (7.68, 9.04, 9.26, 14.1, per 1,000 person years follow-up, respectively, P = .0014). Relative risks for ALL-CVA for each quartile of total and differential WBC counts were obtained using Cox proportional hazards, using the lowest quartile as the reference group. After full adjustment including age, cardiovascular risk factors, fibrinogen, prevalent CHD, cancer or COPD, and aspirin/NSAID use, the relative risks in the highest quartiles of total WBC, neutrophil, and granulocyte counts were 1.63 (95%CI = 1.05–2.54, P = .03), 2.19 (95%CI = 1.41–3.39, P < .001) and 1.91 (95%CI = 1.25–2.92, P = .003), respectively. These significant associations were also seen for TE-CVA, but not for HEM-CVA. No significant associations were found between lymphocyte or monocyte counts and incident stroke or subtypes.
In elderly Japanese-American men, higher total WBC, neutrophil, and granulocyte counts were independent predictors of overall stroke, as well as thromboembolic stroke. Further studies are needed to establish cut-points and treatment options.